Probing the sterol binding site of soybean sterol methyltransferase by site-directed mutagenesis: functional analysis of conserved aromatic amino acids in Region 1.

Soybean sterol methyltransferase (SMT) in the presence of AdoMet catalyzes the transmethylation of the delta24-bond of the sterol side chain to produce phytosterols with a methyl(lene) or ethyl(idene) group at C-24. The function of six aromatic amino acids associated with the putative active center of the SMT, i.e., Region 1 that extends from Phe82 to Phe93 in soybean SMT, was studied by site-directed mutagenesis and heterologous expression in BL21(DE3) bacterial cells. The enzyme-generated products were characterized kinetically and by GC-MS analysis. Substitution of the aromatic amino acids at positions 82, 83, 85, 87, 91, and 93 with a leucine residue produced mutant SMTs with varying activities. The mutants converted cycloartenol to 24(28)-methylene cycloartanol [C1-activity] from a few percent to as much as 95% of the control activity. In contrast, none of the leucine mutants were found to catalyze 24(28)-methylene lophenol [C2-activity], suggesting a loss of function associated with the second C1-transfer activity. In contrast to the loss of the second C1-transfer activity of the Phe82Leu, replacement of the Phe82 residue to isoleucine had minimal effect on the first or second C1-transfer activities, suggesting that the increased bulk (branching) in the leucine side chain contributes to significant perturbations in the active site that generate inaccurate positioning of the substrate side chain disfavoring the second C1-transfer activity. Replacement of Tyr83 to phenylalanine resulted in an increase of the specificity constant (kcat/Km) for the substrate of the second C1-transfer activity by a factor of 5 compared to control and an increase of delta24(28)Z-ethylidene sterol formation in the 24-ethyl sterol product set, suggesting that loss of steric bulk from the phenolic hydroxyl group on tyrosine generates a less precise fit of the delta24(28) sterol side chain into the active site favoring the second C1-transfer activity and prompting reaction channeling during catalysis. Circular dichroism spectra, equilibrium dialysis studies of AdoMet, and chromatographic information of the wild-type and Tyr83 mutants confirmed retention of the overall conformation of the enzyme during the experiments. Together, these findings suggest that the amino acids of Region 1 provide a tight substrate orientation imposed by hydrophobic interactions between the sterol side chain and the SMT active site contacts and control the production and processing of the transmethylation pathways governed by the first and second C1-transfer activities.

Biosynthesis of phytosterols. Kinetic mechanism for the enzymatic C-methylation of sterols.

Cloned soybean sterol methyltransferase was purified from Escherichia coli to gel electrophoretic homogeneity. From initial velocity experiments, catalytic constants for substrates best suited for the first and second C1 transfer activities, cycloartenol and 24(28)-methylenelophenol, were 0.01 and 0.001 s-1, respectively. Two-substrate kinetic analysis using cycloartenol and S-adenosyl-l-methionine (AdoMet) generated an intersecting line pattern characteristic of a ternary complex kinetic mechanism. The high energy intermediate analog 25-azacycloartanol was a noncompetitive inhibitor versus cycloartenol and an uncompetitive inhibitor versus AdoMet. The dead end inhibitor analog cyclolaudenol was competitive versus cycloartenol and uncompetitive versus AdoMet. 24(28)-Methylenecycloartanol and AdoHcy generated competitive and noncompetitive kinetic patterns, respectively, with respect to AdoMet. Therefore, 24(28)-methylenecycloartanol combines with the same enzyme form as does cycloartenol and must be released from the enzyme before AdoHcy. 25-Azacycloartanol inhibited the first and second C1 transfer activities with about equal efficacy (Ki = 45 nm), suggesting that the successive C-methylation of the Delta 24 bond occurs at the same active center. Comparison of the initial velocity data using AdoMet versus [2H3-methyl]AdoMet as substrates tested against saturating amounts of cycloartenol indicated an isotope effect on VCH3/VCD3 close to unity. [25-2H]24(28)-Methylenecycloartanol, [28E-2H]24 (28)-methylenelanosterol, and [28Z-2H]24(28)-methylene lanosterol were prepared and paired with AdoMet or [methyl-3H3]AdoMet to examine the kinetic isotope effects attending the C-28 deprotonation in the enzymatic synthesis of 24-ethyl(idene) sterols. The stereochemical features as well as the observation of isotopically sensitive branching during the second C-methylation suggests that the two methylation steps can proceed by a change in chemical mechanism resulting from differences in sterol structure, concerted versus carbocation; the kinetic mechanism remains the same during the consecutive methylation of the Delta 24 bond.

Purification, characterization and catalytic properties of human sterol 8-isomerase.

CHO 2, encoding human sterol 8-isomerase (hSI), was introduced into plasmids pYX213 or pET23a. The resulting native protein was overexpressed in erg 2 yeast cells and purified to apparent homogeneity. The enzyme exhibited a K (m) of 50 microM and a turnover number of 0.423 s(-1) for zymosterol, an isoelectric point of 7.70, a native molecular mass of 107000 Da and was tetrameric. The structural features of zymosterol provided optimal substrate acceptability. Biomimetic studies of acid-catalysed isomerization of zymosterol resulted in formation of cholest-8(14)-enol, whereas the enzyme-generated product was a Delta(7)-sterol, suggesting absolute stereochemical control of the reaction by hSI. Using (2)H(2)O and either zymosterol or cholesta-7,24-dienol as substrates, the reversibility of the reaction was confirmed by GC-MS of the deuterated products. The positional specific incorporation of deuterium at C-9alpha was established by a combination of (1)H- and (13)C-NMR analyses of the enzyme-generated cholesta-7,24-dienol. Kinetic analyses indicated the reaction equilibrium ( K (eq)=14; DeltaG(o')=-6.5 kJ/mol) for double-bond isomerization favoured the forward direction, Delta(8) to Delta(7). Treatment of hSI with different high-energy intermediate analogues produced the following dissociation constants ( K (i)): emopamil (2 microM)=tamoxifen (1 microM)=tridemorph (1 microM)<25-azacholesterol (21 microM)

Site-Directed Mutagenesis of the Sterol Methyl Transferase Active Site from Saccharomyces cerevisiae Results in Formation of Novel 24-Ethyl Sterols.

Delta(24(28))-Sterols are end products of a mono C-methylation pathway catalyzed by the native Delta(24(25))- to Delta(24(28))-sterol methyl transferase (SMT) enzyme from Saccharomyces cerevisiae. Using a Tyr(81) to Phe mutant SMT enzyme of S. cerevisiae, generated by site-directed mutagenesis of a highly conserved residue in the sterol binding site, we found that several Delta(24(25))- and Delta(24(28))-sterols, which are not substrates for the native protein, were catalyzed to mono- and bis-C24-alkylated side chains. The mutant protein behaved similarly to the native protein in chromatography and in binding zymosterol, the preferred substrate. Zymosterol was converted to fecosterol by the Y81F mutant protein with similar turnover efficiency as the native protein (K(m) = 12 &mgr;M and k(cat) = 0.01 s(-)(1)); trace 24-ethyl sterols were detected from these incubations. 4alpha-Methyl zymosterol, which is not a normal substrate for the wild-type SMT enzyme, was converted to 4alpha-methyl fecosterol in high yield. When fecosterol and 4alpha-methyl fecosterol were assayed individually at saturating concentrations only fecosterol served as an effective substrate for the second C-transfer step (K(m) = 38 &mgr;M and k(cat) = 0.002 s(-)(1)), suggesting that successive C-methylation of Delta(24(28))-substrates is limited by product release and that molecular recognition of sterol features involves hydrogen bond formation. Isomeric 24-ethyl sterol olefins generated from 24(28)-methylene cholesterol were characterized by chromatographic (GC and HPLC) and spectral methods (MS and (1)H NMR), viz., fucosterol, isofucosterol, and clerosterol. Changes in rate of C-methylation and product distributions resulting from deuterium substitution at C28 were used to establish the kinetic isotope effects (KIEs) for the various deprotonations leading to C24-methylene, C24-ethylidene, and C24-ethyl sterols. An isotope effect on C28 methyl deprotonation generated during the first C(1)-transfer was detected with zymosterol and desmosterol paired with AdoMet and [(2)H(3)-methyl]AdoMet. A similar experiment to test for a KIE generated during the second C(1)-transfer reaction with AdoMet paired with 24(28)-methylenecholesterol and [28-(2)H(2)]24(28)-methylene cholesterol indicated an inverse isotope effect associated with C27 deprotonation. Alteration in the proportion of the C24 alkylated olefinic products generated by the pure Y81F mutant resulted from the suppression of the formation of Delta(24(28))-ethylidene sterols (C28 deprotonation) by a primary deuterium isotope effect with a compensating stimulation of the formation of 24-ethyl sterols (C27 deprotonation). Kinetic study on the rate of product formation indicated a normal KIE of k(H)/k(D) = 2.62 for the first C(1)-transfer. Alternatively, an inverse KIE was established with k(H)/k(D) = 0.9 for the second C(1)-transfer resulting from conversion of the 24(28)-double bond (sp(2) hybridization) to a 24beta-ethyl group (sp(3) hybridization). From the structures and stereochemical assignments of the C-ethyl olefin products, the stereochemistry of the attack of AdoMet in the second C(1)-transfer was found to operate a Si-face (backside) attack at C24, analogous to the first C(1)-transfer reaction.